Questions under study: To assess the diabetes-related knowledge of medical and nursing house staff with particular focus on inpatient diabetes management and insulin therapy. Methods: In a cross-sectional design, diabetes-related knowledge among physicians, graduate nurses, medical students and student nurses of the departments of internal medicine, surgery and gynaecology was assessed using a 42-item multiple-choice questionnaire. Results: Of 466 subjects approached 232 completed the questionnaire. Knowledge was highest for physicians in internal medicine (total score 62 +/- 11%, mean +/- SD, max. 100%). Physicians in surgery and gynaecology had lower scores (48 +/- 14% and 47 +/- 12%, p <0.001 and p <0.05 respectively, compared with internal medicine), which were comparable to those of medical students (49 +/- 9%). Knowledge of attending physicians and residents did not differ within the three specialities. Nurses in internal medicine and surgery had the same level of knowledge (total score 41 +/- 11% each), which was comparable to that of student nurses (40 +/- 9%). Nurses in gynaecology had lower total scores (30 +/- 10%, p <0.001 compared with nurses in medicine and surgery respectively). The comfort level in dealing with diabetes correlated with the knowledge score for physicians, but not for nurses. Conclusions: Knowledge of diabetes is mediocre among medical and nursing house staff. For physicians, the knowledge level depends on the speciality (internal medicine better than surgery and gynaecology) and correlates with the comfort level in dealing with diabetes, but is not higher in attending physicians than in residents. Nurses have a high comfort level, which, however, does not correlate with knowledge, which is similar in medicine, surgery and student nurses, but low in gynaecology.